|Home | About | Journals | Submit | Contact Us | Français|
Catalytic subunits of PI3K play a critical role in growth factor signaling and survival by phosphorylating inositol lipids. We found that PI3K Class IA p110α and p110β have distinct functions in myoblasts. Inhibition of p110α reduced IGF-I-stimulated Akt activity and prevented IGF-I-mediated survival in H2O2-treated cells; in contrast, siRNA knockdown of p110β increased IGF-I-stimulated Akt activity. However, inhibition of p110β catalytic activity did not increase IGF-I-stimulated Akt activity, suggesting a role for p110β protein interactions rather than decreased generation of phosphoinositides in this effect. Increased Akt activity in p110β-deficient myoblasts was associated with diminished ERK activation as well as ERK-dependent IRS-1 636/639 phosphorylation, findings we show to be independent of p110β catalytic function, but associated with IGF-IR endocytosis. We also report that IGF-I protects myoblasts from H2O2-induced apoptosis through a mechanism that requires p110α, but may be independent of Akt or ERK under conditions of Akt and ERK inhibition. These observations suggest that both p110α and p110β are essential for growth and metabolism in myoblasts. Overall, our results provide new evidence for the roles of p110 isoforms in promoting cellular proliferation and homeostasis, IGF-IR internalization, and in opposing apoptosis.
Members of the phosphoinositide-3-OH kinase (PI3K) class of enzymes generate phosphoinositol products that act as second-messengers in a number of intracellular signaling cascades (1). PI3K catalytic enzymes are categorized into three classes by their structure, substrate specificity, and lipid products (2). Members of the Class IA PI3Ks (α, β, and δ) are heterodimers consisting of a 110 kDa catalytic subunit and an 85, 55, or 50 kDa regulatory subunit (2, 3). Activation of PI3K leads to proliferation and pro-survival effects; however, dysregulation of PI3K signaling can promote aberrant proliferative signals leading to cellular transformation (4-7).
Examination of the ubiquitously expressed PI3K Class IA p110α and p110β catalytic subunits has revealed distinct and redundant roles for each depending on cellular context (8-10). Heterozygous knock-in of catalytically-inactive p110α or pharmacological inhibition of p110α with a small-molecule inhibitor has been shown to negatively regulate growth, metabolism, and growth factor signaling in mice (10, 11). p110β has been shown to mediate cell growth and development, DNA replication, insulin sensitivity, tumorigenesis, and G-protein-coupled receptor-activated signaling (12-15). Recently, a kinase-independent, scaffolding role of p110β has been suggested; indeed, this kinase-independent function may play a role in cell proliferation and clatharin-mediated endocytosis (13, 16). Mice doubly heterozygous for p110α and p110β showed mild glucose intolerance and reduced sensitivity to insulin challenge associated with p85 protein instability, but showed no difference in insulin-stimulated Akt phosphorylation or activity in liver or muscle as compared to control mice (17).
Insulin-like growth factor I (IGF-I) is a hormone that promotes proliferation, differentiation, and survival in a number of cells types mediated principally through PI3K/Akt, and Ras/Raf/MEK/ERK pathways (18-20). IGF-I/PI3K signaling has been shown to promote survival in several apoptosis-inducing models such as serum withdrawal and oxidative stress (21-23); however, the specific PI3K and Akt isoforms involved in survival have yet to be established. Likewise, PI3K/Akt-independent compensatory survival mechanisms have yet to be fully resolved. We report here that C2C12 myoblasts transfected with siRNA against p110α, p110β, or a combination of both, displayed differential phenotypes with respect to cell growth and PI3K-dependent signaling. Furthermore, we found that IGF-I stimulation differentially regulated Akt phosphorylation and activation in a PI3K- and Akt- isoform-specific fashion, and that p110α was the principal Class IA PI3K mediating IGF-I anti-apoptotic actions. Finally, we report that knockdown p110β negatively regulates IGF-IR internalization and ERK activation, which was associated with relief of feedback inhibition of PI3K-dependent signal transduction.
We first determined that both p110α and p110β are expressed at the mRNA and protein level in C2C12 myoblasts (Figure 1a), but that p110γ and p110δ were not expressed appreciably. In order to establish efficacy of siRNA-mediated knockdown of p110α and p110β, cells were transfected with siRNA against p110α (si-p110α), p110β (si-p110β), or non-targeting control siRNA (si-Con). Addition of si-p110α resulted in ~60-65% reduction in p110α mRNA, while addition of si-p110β resulted in ~65% reduction in p110β mRNA (Figure 1b). mRNA levels of p110β were unaffected by p110α knockdown, and mRNA levels of p110α were unaffected by p110β knockdown. To determine whether cell growth was affected by knockdown of p110α, p110β, or the combination (si-p110α + β), cells were counted at 24 and 48 hours after transfection. Addition of si-p110β and si-p110α + β caused a reduction in cell number at 48-hours for si-p110β (P<0.05) and at 24- and 48-hours for si-p110α + β (Figure 1c). This reduction in cell number was associated with increased cleavage of Caspase-3 and Poly (ADP)-ribosyl-polymerase (PARP), a molecule downstream of Caspase-3 and an indicator of apoptosis (Figure 1d)(24). Because knockdown of p110α can reduce PIP(3,4,5)3 levels (11), and p110β may act to maintain basal levels of PIP(3,4,5)3 (10), we analyzed Akt (also known as protein kinase B), an AGC kinase family members whose activation is known to be regulated by phosphoinositides (25). Forty-eight hours after transfection, levels of phospho-Akt in the hydrophobic domain (Ser473/474/472 in Akt1, Akt2, and Akt3, respectively; hereafter referred to as S473) showed decreased phosphorylation after treatment with p110α; conversely, phospho-Akt Ser 473 levels were increased after treatment with si-p110β alone (Figure 1e). The levels of phosphorylated Akt did not correlate with total p85 levels, as reductions in total p85 were observed in p110α- and p110β- deficient cells, but were even further reduced when cells were treated simultaneously with si-p110α and si-p110β (Figure 1e).
Because IGF-I can mediate pro-survival effects through Akt, we sought to determine the responsiveness of p110α- and p110β-deficient cells to IGF-I. Addition of IGF-I to cells transfected with either si-p110α, si-p110β, or si-p110α + β resulted in increased Akt phosphorylation at the activation loop (Thr308/309/305 in Akt1, Akt2, and Akt3, respectively; hereafter referred to as T308). This increase was reduced in si-p110α-treated cells, but further increased in si-p110β-transfected cells (Figure 2a). IGF-I-stimulated phospho-Akt in cells treated with both si-p110α and p110β together was not statistically lower than si-Con cells treated with IGF-I (Figure 2a). Because the sequences surrounding the Akt active site (T308) and hydrophobic motif (S473) residues are similar between isoforms, and because the antibody cannot discriminate between isozymes, we performed in vitro kinase assays to determine relative contributions of each isoform. IGF-I-stimulated Akt1 activity was reduced in cells transfected with si-p110α and si-p110α + β; conversely, Akt1 activity was increased in response to IGF-I in cells treated with si-p110β (Figure 2b). Only cells transfected with si-p110α alone showed reduced Akt3 activity, and there was no increase or decrease in Akt3 activity in cells transfected with either p110β or p110α + β, respectively. Akt2 activity was unchanged by any treatment, and long exposures (shown in Figure 2b) likely represent background. Taken together, these data suggest that knockdown of p110α reduces IGF-I-stimulated Akt phosphorylation and activation of both Akt1 and Akt3. IGF-I-induced Akt1 activity was greater in si-p110β-transfected cells than IGF-I-stimulated Akt1 activity in control cells, which correlated with increased Akt phosphorylation. Yet, IGF-I-stimulated Akt3 activity did not show this elevation, suggesting that the increase in total Akt phosphorylation may be the result of increased Akt1 activation in p110β-deficient cells. On the other hand, in cells transfected with si-p110α + β, Akt1 activity was diminished, but Akt3 activity was not, suggesting that the increase in IGF-I-stimulated Akt phosphorylation in si-p110α + β-treated cells as compared to si-p110α alone, reflects increased Akt3 activation.
To test whether this increased Akt activity in p110β knockdown cells correlated with increased p110 association with components of the IGF-IR signaling complex, we performed immunoprecipitation experiments. Treatment of cells with IGF-I promoted p110α and p110β association with IRS-1 and IGF-IR at early time points following administration (Figure 3a, left). In p110α-deficient cells, IGF-I-induced association of both p110α and p110β with IGF-IR and IRS-1 was reduced, whereas in p110β-deficient cells, association of p110α with IGF-IR and IRS-1 was enhanced (Figure 3a, right). Simultaneous knockdown of p110α and p110β prevented IGF-I induced association of both p110α and p110β with IGF-IR and IRS-1. These findings suggest that IGF-I-induced association of p110α with IGF-IR and IRS-1 is enhanced when p110β levels are reduced. We next determined whether p110α or p110β deficiency affected IGF-I-stimulated PI(3,4,5)P3 production. IGF-I promoted PI(3,4,5)P3 generation from p110α and p110β immunoprecipitates, and this effect was reduced when cells were deficient in the respective isoform (Figure 3b). Levels of PI(3,4,5)P3 in IGF-I-treated cells transfected with si-p110β were intermediate between unstimulated and IGF-I-stimulated control cells. Altogether, these data suggest that Akt activation correlates with levels of p110/IGF-IR complex, but in cells deficient in p110β, mechanisms other than PI(3,4,5)P3 production may contribute to elevated Akt activation.
To further elucidate the mechanism whereby Akt phosphorylation and activity is increased in p110β-deficient cells, as well as to further characterize intracellular responses to p110 catalytic subunit deficiency, we examined other molecules involved in canonical IGF-I signaling pathway. In cells deficient in p110β alone, or deficient in p110α and p110β, we found that phosphorylation of ERK was reduced in the basal state as well as after stimulation with IGF-I (Figure 4a). Additionally, IGF-I-stimulated phosphorylation of IRS-1 as Ser 636/639 was attenuated under conditions of reduced p110β, a finding consistent with previous work suggesting a dependence on ERK for IRS serine phosphorylation at this site (26).
To determine whether decreased ERK phosphorylation seen in p110β-deficient cells was due to an overall decrease in p110β protein levels or due to decreased catalytic activity, cells were treated with increasing concentrations of TGX-221, an inhibitor of p110β catalytic function. Addition of TGX-221 reduced LPA-induced Akt phosphorylation in a dose-dependent fashion, but did not affect IGF-I-stimulated ERK phosphorylation at any concentration tested (Figure 4b). These data suggest that the decreased levels of phosphorylated ERK observed in p110β-deficient cells may result from a loss of total p110β rather than loss of catalytic function, a finding consistent with a scaffolding action of p110β.
To address whether this catalytic-independent function of p110β mediates IGF-IR internalization, we examined the presence of IGF-IR at the cell surface after IGF-I treatment in the absence or presence of TGX-221 or si-p110β. Treatment of myoblasts with IGF-I resulted in a reduction in cell-surface IGF-IR that was maximal after 30-minutes (Figure 4c, left). In IGF-I-stimulated cells pre-treated with 100 nM TGX-221, IGF-IR internalization was similar to IGF-I treatment alone; however, in cells deficient in p110β, IGF-IR did not internalize in response to IGF-I (Figure 4c, right). These findings suggest that at least one kinase-independent role of p110β in myoblasts is to mediate IGF-IR internalization.
To establish whether the differential signaling effects of p110α and p110β knockdown extend to physiological actions in a pro-apoptotic milieu, cells were examined to determine whether knockdown of p110 isoforms was sufficient to inhibit IGF-I from preventing oxidative stress-induced cell death. To first test this, we employed an inhibitor specific to PI3K p110α (“p110αi”) (27), since broad-spectrum PI3K pharmacological inhibitors such as LY294002 and wortmannin do not discriminate between isoforms. Caspase-3 and PARP cleavage were increased in cells treated with 400μM H2O2 for 4-hours, an effect that was completely prevented by a 30-minute pre-treatment with IGF-I (Figure 5a and 5b, lanes 1-4). Caspase-3 and PARP cleavage was increased in p110αi and p110αi + H2O2 -treated cells, and p110αi completely prevented IGF-I from inhibiting apoptosis (Figure 5a and 5b lanes 7-8; compare to lanes 1-2). In contrast, treatment of myoblasts with IGF-I in the presence of the p110β inhibitor TGX-221 did not prevent IGF-I from reducing H2O2-stimulated Caspase-3 and PARP cleavage (Figure 5c, right).
To confirm these results, we employed siRNA directed against p110α and p110β. Cells were pre-treated with IGF-I, treated with H2O2, or both in the presence or absence of si-p110α, si-p110β, or si-p110α + β. Similar to the inhibitor results, IGF-I was unable to prevent H2O2-induced cleavage of caspase-3 and PARP in cells deficient in p110α whether transfected with si-p110α alone (Figure 5d left) or in combination with si-p110β (Figure 5d, right). si-p110β-treated cells appeared partially resistant to H2O2-induced apoptosis (Figure 5d, middle), possibly a result of increased Akt activation (Figure 2). Taken together, these data suggest that IGF-I acts through PI3-K p110α isoform to prevent H2O2-induced apoptosis in myoblasts exposed to oxidative stress.
Since reduction of p110β effectively inhibited phosphorylated ERK levels while simultaneously increasing Akt activation, we next tested whether Akt was a primary survival intermediate in p110β knockdown cells. Maximal inhibition of IGF-I-stimulated phosphorylation of Akt S473 was obtained at a concentration of 10 μM using a compound specific for Akt (“Akti;” ref. (28)) (Figure 5e, left). IGF-I was unable to prevent H2O2-induced apoptosis in p110β-transfected cells (Figure 5e, right) also treated with Akti, thus confirming that a functional Akt pathway is necessary for survival in p110β-deficient myoblasts.
We noted that knockdown of p110α or p110α + β increased Caspase-3 and PARP cleavage (Lanes 1-2 compared to lanes 5-6 Figure 5a, 5b, and 5d), an effect that was reversed by IGF-I administration. To determine the identity of pathway(s) or molecule(s) involved in this effect, cells were treated with an inhibitor of Akt (Akti) or of MEK (U0126). Cells were pre-incubated with inhibitors of increasing concentrations for one hour, and then exposed to IGF-I for 4.5 hours. Ten micromolar U0126 completely prevented IGF-I-stimulated phosphorylation of p42/p44 ERK at 30-minutes, and 4.5-hours after IGF-I administration, levels of phospho-ERK remained slightly below that of IGF-I-naive cells (Figure 6a, left). Cells treated with si-p110α or si-p110α + β were pre-incubated with either Akti or U0126 and then stimulated with IGF-I for 4.5 hours. Pharmacological blockade of MEK in IGF-I-treated p110α-deficient cells prevented IGF-I-induced survival, and blockade of Akt increased p110α-deficiency induced apoptosis beyond that of the MEK inhibitor (Figure 6a, right). However, in p110α + p110β-deficient cells, inhibition of Akt, but not MEK, fully prevented IGF-I pro-survival signaling. These data suggest that when p110α, and by extension Akt activation, is reduced, IGF-I survival signaling compensates by increased flux through MAPK pathway. On the other hand, when cells are deficient in both p110α and p110β, IGF-I signals primarily through Akt to promote survival.
Because IGF-I differentially acts through Akt and MEK pathways to promote survival from p110α and p110α + p110β- deficiency-induced apoptosis, we sought to determine whether these pathways were also involved in preventing H2O2-induced apoptosis. Blockade of either Akt, ERK, or the concurrent inhibition of both, did not prevent IGF-I from inhibiting Caspase-3 or PARP cleavage (Figure 6b). These data suggest that, although Akt and ERK pathways can compensate for each other in response to IGF-I, at least one more mechanism exist through which IGF-I can exert anti-apoptotic effects in response to oxidative stress in myoblasts.
In this work, we report that myoblasts deficient in p110α and/or p110β show disparate growth and intracellular signaling phenotypes, corresponding with differential regulation of Akt and ERK signaling pathways. One key observation was that knockdown of p110α inhibited, but knockdown of p110β promoted, the activation of Akt in an isoform-specific manner. Although p110β may play a secondary role in growth factor signaling by maintaining a basal pool of PI (3, 4, 5)P3 (10, 13), it is not generally thought to play a primary role in RTK-instigated signaling except under certain conditions such as PTEN loss, or in certain cell types (2, 29). Our results suggest that p110β deficiency leads to increased Akt activation through two separate mechanisms: First, reduced p110β levels enhance the IGF-I-induced association of p110α with IGF-IR and IRS-1. Second, reduced p110β levels inhibit phosphorylation of ERK and IRS-1 at S636/639, thereby relieving IRS-associated negative feedback. One intriguing finding was that increased IGF-I-stimulated Akt activation in p110β-deficient cells did not correlate with increased PI(3,4,5)P3 production, a finding consistent with previous results in skeletal muscle of insulin-stimulated p110α +/- /p110β +/- mice (17). Those authors suggested that decreased p85 levels may contribute to insulin-sensitivity, and indeed, in our study we observed lowered p85 in p110β-deficient cells (Figure 1e).
Furthermore, our finding that serine phosphorylation of IRS-1 at S636/639, a site dependent on ERK (26) is reduced, is consistent with a relief of negative PI3K-signaling feedback (30). Since IGF-I-stimulated Akt activity did not statistically differ between control and p110α + p110β-deficient cells (Figure 2b), yet there was virtually no p110/IGF-IR/IRS-1 association (Figure 2c, right), the loss of feedback inhibition in p110α + p110β-deficient myoblasts may explain the mechanism underlying the similarity in Akt activation. Indeed, in skeletal muscle of insulin-stimulated p110α +/- /p110β +/- mice, Akt phosphorylation was similar to that observed in muscle of insulin-treated wild type mice. It is possible that other mechanism(s) contribute to increased Akt activation in p110β-deficient cells, including decreased actions of Akt-specific phosphatases (31, 32), but whether activation of these phosphatases is reduced under these conditions remains to be investigated. Altogether, our observations suggest that increased IGF-I-stimulated Akt activity in p110β-deficient myoblasts may result from both an increase in p110α/IRS-1/IGF-IR association as well as decreased negative feedback to IRS-1 secondary to reduced ERK activation. That distinct Akt isoforms are differentially regulated suggests a high degree of fine-tuning in these processes. Figure 7 presents this signaling model diagrammatically.
The lower levels of ERK phosphorylation in p110β deficient cells may be reflective of decreased association of p110β with Ras, an upstream molecule in the MAPK signaling pathway; indeed, ERK activation has been shown to be attenuated under conditions where p110β cannot bind to Ras (4, 5). However, use of TGX-221, a compound that inhibits p110β catalytic activity, did not prevent ERK phosphorylation in response to IGF-I in this study. These results suggest that non-catalytic functions of p110β may be essential for ERK activation in myoblasts; in support of this, cells deficient in p110β also showed decreased IGF-IR internalization in addition to attenuated ERK signaling. This is in agreement with previous work suggesting a requirement for IGF-IR internalization for activation of upstream MAPK pathway components in CHO cells (33); additionally, p110β itself has been shown to modulate EGFR and transferrin uptake (13, 16). Taken together, these data and ours support the hypothesis of a kinase-independent function of p110β in IGF-IR internalization and signaling. Consequently, p110β may play a role in IGF-I-mediated processes in skeletal muscle such as progenitor cell proliferation, survival, and differentiation, which occur during muscle development and repair/regeneration.
In light of the findings above, it was not surprising that inhibition of p110α by means of pharmacological blockade or siRNA completely prevented IGF-I from reducing Caspase-3 and PARP cleavage in response to H2O2, whereas TGX-221- or RNAi-induced silencing of p110β did not (Figure 5a-d). Akt is a critical mediator of anti-apoptotic signaling, acting to suppress apoptosis by phosphorylation and inactivation of pro-appoptotic molecules such as apoptosis-signal regulating kinase-1 (Ask1) (34), Bad (35), and FoxO3a (36). Because Akt signaling was reduced after p110α knockdown or treatment with p110αi, it would logically follow that suppression of Akt-targeted pro-apoptotic molecules would also be attenuated. Likewise, because Akt activation was elevated with p110β knockdown, and this was found to be necessary for survival, IGF-I-stimulated Akt actions would also be elevated.
Myoblasts deficient in p110α or p110α and p110β together showed increased Caspase-3 and PARP cleavage, an effect fully reversed by IGF-I. We found that inhibition of MEK prevented IGF-I from attenuating p110α-deficiency-induced Caspase-3 and PARP cleavage, and that inhibition of Akt caused apoptosis in excess of p110α deficiency alone. Inhibition of Akt, but not MEK, prevented IGF-I from attenuating p110α + p110β-deficiency-induced Caspase-3 and PARP cleavage (Figure 6a). These data suggest that IGF-I can act through a MEK-dependent pathway to rescue cells from apoptosis associated with p110α deficiency, but that myoblasts under these conditions are more sensitive to loss of Akt, possibly resulting from the reduced Akt activation associated with p110α deficiency. Likewise, in cells simultaneously deficient in p110α and p110β, blockade of Akt prevented IGF-I from opposing apoptosis, whereas blockade of MEK had minimal effects. This is consistent with our findings that ERK activation is inherently reduced under conditions of p110β deficiency, whether alone or in combination with p110α reduction.
We found that IGF-I was still able to prevent H2O2 -induced apoptosis even under conditions of Akt and MEK blockade. However, our results also suggest a dependence on p110α. Therefore, our data collectively suggest that IGF-I is capable of preventing H2O2-induced apoptosis through an Akt- and MEK-independent, but p110α-dependent mechanism. Since other AGC kinases are also regulated through p110α in response to IGF-I, it is possible that one or several may contribute to opposing apoptosis. For example, serum- and glucocorticoid-inducible kinase 1 (SGK1) can phosphorylate and inhibit pro-apoptotic FoxO3a at the same sites as Akt, although there exists a preference for some sites over others (37).
In conclusion, we found that knockdown of p110α and p110β have differential effects on cell growth, survival signaling, and Akt activation. We have defined a kinase-independent role for p110β in IGF-IR internalization and ERK phosphorylation, which is associated with negative feedback to the IGF-IR/IRS-1 signaling complex. Additionally, we identified p110α as the primary Class IA PI3K involved in IGF-I-mediated rescue from oxidative stress-induced apoptosis and that, in the absence of Akt and MEK-dependent signaling, IGF-I can still prevent H2O2-induced apoptosis, but through a mechanism that requires p110α.
C2C12 murine myoblasts were purchased from ATCC (Manassas, VA). H2O2 and bovine serum albumin (fraction V) was purchased from Sigma (St. Loius, MO). rhIGF-I was purchased form Austral Biologicals (San Ramon, CA). Primary antibodies directed against p110β and GAPDH and HRP-linked secondary antibodies were purchased from Santa Cruz (Santa Cruz, CA); all other antibodies were obtained from Cell Signaling Technologies (Danvers, MA). PI3Kα inhibitor IV (“p110αi” in this manuscript, corresponds to figure 15e in ref (27)), TGX-221, Akt 1/2 inhibitor VIII (“Akti”), and U0126 were purchased from Calbiochem (San Diego, CA). LPA (18:1 oleoyl) was purchased from Avanti Polar Lipids (Alabaster, AL).
C2C12 myoblasts were maintained in high-glucose DMEM containing 10% FBS and antibiotics (growth medium), with medium being replenished after 24-hours. Forty-eight hours after initial seeding cells were ~95% confluent and unless otherwise indicated, experiments were conducted under these conditions (in growth medium). Pre-designed Silencer Select siRNAs for mouse p110α (ID# s71604), p110β (ID# s93108), and negative controls (non-targeting siRNA ID# 4390843) were purchased from Ambion Inc. (Austin, TX). Cells (1.8 × 105/well in 6-well culture dish) were reverse-transfected with double-stranded siRNA in antibiotic-free DMEM plus 10% FBS using Lipofectamine 2000 according to manufacturer's instructions (Invitrogen). Twenty-four hours after reverse transfection, medium was changed to DMEM with 10% FBS containing antibiotics. RNA was isolated 48-hours after transfections using RNA-STAT (Tel-test, Friendswood, TX), quantified, and the level of silencing quantitated by real-time PCR as described later in this section. In some cases, protein was isolated 48h after transfection and the level of silencing determined by Western Blot.
Medium was removed from plates and monolayers detached with 0.05% Trypsin-EDTA (Gibco). Cells were suspended in growth medium and counted in a hemacytometer chamber.
Conversion of total RNA to single-strand cDNA was accomplished using the High-Capacity cDNA Archive Kit (P/N 4322171; Applied Biosystems, Foster City, CA). Briefly, 2 μg total RNA was reverse transcribed using random primers for the following incubation times: 25° C for 10 minutes, then 37° C for 2 hours. cDNA samples were stored at -80° C until use. TaqMan-MGB p110α (Mm00435673_m1), p110β (Mm00659576_m1), p110γ (Mm00445038_m1), p110δ (Mm00435674_m1) and B2M (Mm00437762_m1) probe and primers were purchased from Applied Biosystems as “Gene Expression Assays.” The real-time PCR reaction was performed within an ABI 7500 thermal cycler. The fluorescence of 3 to 15 cycles was set up as background. Data was collected at the annealing step of each cycle, and the threshold cycle (Ct) for each sample calculated by determining the point at which the fluorescence exceeded the threshold limit. The standard curve was calculated automatically via software by plotting the Ct values against each standard of known concentration and calculation of the linear regression line of this curve. Serially diluted amounts of RNA were used to establish standard curves. All samples were run in duplicate.
C2C12 cells were harvested in ice-cold lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% Triton, 2.5 mM sodium pyrophosphate, 1 mM beta-glycerophosphate, 1 mM Na3VO4, 1 μg/ml Leupeptin, 1 mM PMSF, and 1: 100 dilution of phosphatase inhibitor cocktails 1 and 2 (Sigma)). Homogenates were triturated through a small-bore needle, incubated on ice for 30 min, and then centrifuged at 14,000 × g for 10 min at 4°C. Protein concentrations were determined by the method of Bradford (38). For immunoprecipitations, 200ug of protein lysate were co-incubated with 4uL of respective antibody in 200uL cell lysis buffer with gentle rotation overnight at 4°C. 20uL of Protein A/G agarose slurry (Santa Cruz) were then added and rotated an additional 2 - 4 hours at 4°C. Complexes were centrifuged at 14, 000 × g and washed 5 times in wash buffer before addition of 20uL 2× Laemmli buffer and then boiled for 5 minutes. For Western blotting, equal amounts of cell lysate proteins (typically 25μg) were electrophoresed through denaturing SDS-PAGE. Proteins were transferred to PVDF membranes (Millipore Corp., Bedford MA). Membranes were incubated for 1 h in 5% dry milk solution in Tris-buffered saline plus 0.5% Tween-20 (TBST) and then incubated with the appropriate primary antibody at an appropriate dilution (as described in Table 1) overnight in 5% BSA in TBST. Membranes were washed three times in TBST followed by incubation with the appropriate secondary antibody and again washed three times. Membranes were incubated with enhanced chemiluminescence reagents (Thermo Fisher, Rockford, Il) and exposed to film.
Immune complex-kinase assays were performed by a protocol from an assay kit purchased from Cell Signaling technologies. Cell lysates (200 μg) were incubated with anti-Akt1, anti-Akt2, or anti-Akt3 antibody overnight at 4°C in cell lysis buffer. Immune complexes were then rotated for 2 – 4 h at 4°C with protein A-conjugated agarose beads (20 μl of 50% slurry/reaction) before being washed twice in cell lysis buffer and twice in kinase buffer (25 mM Tris-HCl (pH 7.5), 5 mM beta-glycerophosphate, 2 mM dithiothreitol, 0.1 mM Na3VO4, 10 mM MgCl2). After resuspension in kinase assay buffer containing ATP and a GST-GSK-3 α/β fusion protein (residues surrounding GSK-3 α/β Ser21/9 (CGPKGPGRRGRRRTSSFAEG)), the reaction was allowed to proceed at 30°C for 30 min. After the reaction was stopped by addition of concentrated SDS-PAGE loading buffer, samples were separated by SDS-PAGE and transferred to PVDF membranes as described above. Immunoblotting was performed using primary antibodies to phospho-GSK-3α/β provided in the kinase assay kit, followed by addition of HRP-conjugated secondary antibodies, detection by ECL, and exposure to X-ray film. Results were quantitated by densitometry.
Analysis of IGF-IR at the cell surface was accomplished through use of a commercially available cell surface protein isolation kit (Pierce). Briefly, for each treatment, four 100mm culture dishes were seeded with C2C12 cells and transfected as described above. Cells were labeled with Sulfo-NHS-SS-biotin for 30-minutes and the reaction quenched with the provided quenching solution. Cells were transferred to 50 mL conical tubes, centrifuged, and washed with TBS. Lysis buffer provided in the kit was added to cell pellets and transferred to microfuged tubes and briefly sonicated every 10-minutes for 30-minutes at a power of 1.5 followed by clarification of supernatant which was quantified and 1.4 g added to provided spin columns. At this point, some protein lysate was stored at -80° C for analysis of total IGF-IR or GAPDH. Biotin-labeled proteins were washed and isolated with NeutrAvidin Agarose (provided in kit) with end-over-end mixing for one hour. Agarose/biotinylated complexes were washed and incubated in sample buffer (62.5 mM Tris, pH 6.8, 2% SDS, 20% glycerol) for one hour, centrifuged, and the eluate collected. A small amount of bromophenol blue was added to eluate and proteins were separated by SDS-PAGE followed by Western blotting for IGF-IR or GAPDH as described above.
Levels of PI(3,4,5)P3 were determined by using a commercially-available ELISA assay (Echelon Biosciences, Salt lake City, UT; P/N K-1000) following the manufacturers instructions. Briefly, after treatment, cells were washed three times in Buffer A (20 mM Tris-HCl, pH 7.4, 137 mM NaCl, 1 mM CaCl2, 1 mM MgCl2, and 0.1 mM sodium orthovanadate) and incubated at 4-degrees C for 30-minutes in Buffer A plus 1% NP-40 and 1 mM PMSF. Insoluble material was removed by centrifugation and clarified lysates quantified. 125 μg was immunoprecipitated with either p110α or p110β antibody for 2-hours at 4-degrees C, followed by mixing with protein A/G agarose beads (Santa Cruz) for 1-hour. Immunocomplexes were washed three times with Buffer A plus 1% NP-40, three times with 0.1 M Tris-HCl, pH 7.4; 5 mM LiCl, and 0.1 mM sodium orthovanadate, and twice in TNE (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, and 0.1 mM sodium orthovanadate). Kinase reactions were performed in triplicate by incubation in PI3K reaction buffer (20 mM Tris, pH 7.4, 10 mM NaCl, 4 mM MgCl2, and 25 μM ATP) and 100 μM PI(4,5)P2 substrate for 2-hours, followed by centrifugation to stop the reactions and addition of PI(3,4,5)P3 detector for 1-hour. At this time, PIP3 standards (provided) were set up using concentrations of 200, 100, 50, 25, 12.5, and no lipid control. Reacted mixtures and standards were then transferred to the detection plate for 60-minutes, washed once in TBS-T, and secondary detector added for 30-minutes. Wells were washed three times with TBS-T and then 100 μl of TMB solution added to induce color change. Color development was stopped by addition of 1 N H2SO4 stop solution when color development was within the linear range of standards, followed by reading at 450 nM.
Data are presented as means ± S.E.M. Statistics were performed using either one-way or two-way ANOVA and multiple range tests a posteriori, as described in figure legends. Densiometric analysis was performed using Image J 1.60 (NIH). A P-value <0.05 was considered significant.
The authors wish to thank Dr. John C. Lee of the Department of Biochemistry, University of Texas Health Science Center at San Antonio, for many helpful discussions during the preparation of this work.
This work was supported by NIA grant R01AG026012 to MLA. RWM was supported by pre-doctoral award from NIA training grant T32 AG021890-08.